Hidden Physics Models: Machine Learning of Nonlinear Partial Differential Equations

Abstract

While there is currently a lot of enthusiasm about "big data", useful data is usually "small" and expensive to acquire. In this paper, we present a new paradigm of learning partial differential equations from {\em small} data. In particular, we introduce \emph{hidden physics models}, which are essentially data-efficient learning machines capable of leveraging the underlying laws of physics, expressed by time dependent and nonlinear partial differential equations, to extract patterns from high-dimensional data generated from experiments. The proposed methodology may be applied to the problem of learning, system identification, or data-driven discovery of partial differential equations. Our framework relies on Gaussian processes, a powerful tool for probabilistic inference over functions, that enables us to strike a balance between model complexity and data fitting. The effectiveness of the proposed approach is demonstrated through a variety of canonical problems, spanning a number of scientific domains, including the Navier-Stokes, Schrödinger, Kuramoto-Sivashinsky, and time dependent linear fractional equations. The methodology provides a promising new direction for harnessing the long-standing developments of classical methods in applied mathematics and mathematical physics to design learning machines with the ability to operate in complex domains without requiring large quantities of data.

References in corpus (3)

• Applied Koopmanism
• Inferring solutions of differential equations using noisy multi-fidelity data
• Numerical Gaussian Processes for Time-dependent and Non-linear Partial Differential Equations

Cited by in corpus (62)

• Integrating Machine Learning and Multiscale Modeling: Perspectives, Challenges, and Opportunities in the Biological, Biomedical, and Behavioral Sciences
• Prediction of Aerodynamic Flow Fields Using Convolutional Neural Networks
• Variational Physics-Informed Neural Networks For Solving Partial Differential Equations
• A neural network approach for the blind deconvolution of turbulent flows
• On closures for reduced order models $-$ A spectrum of first-principle to machine-learned avenues
• Using Noisy or Incomplete Data to Discover Models of Spatiotemporal Dynamics
• A Physics-Informed Deep Learning Paradigm for Traffic State and Fundamental Diagram Estimation
• Autonomous inference of complex network dynamics from incomplete and noisy data
• Deep Lagrangian Networks: Using Physics as Model Prior for Deep Learning
• Feature engineering and symbolic regression methods for detecting hidden physics from sparse sensors
• Variational Onsager Neural Networks (VONNs): A thermodynamics-based variational learning strategy for non-equilibrium PDEs
• Long short-term memory embedded nudging schemes for nonlinear data assimilation of geophysical flows
• Combining machine learning and data assimilation to forecast dynamical systems from noisy partial observations
• Mesoscopic simulations at the physics-chemistry-biology interface
• A priori analysis on deep learning of subgrid-scale parameterizations for Kraichnan turbulence
• Differentiable Physics-informed Graph Networks
• Probabilistic learning on manifolds constrained by nonlinear partial differential equations for small datasets
• PySINDy: A Python package for the Sparse Identification of Nonlinear Dynamics from Data
• Data-driven discovery of free-form governing differential equations
• A Levenberg-Marquardt algorithm for sparse identification of dynamical systems
• Data-driven, multi-moment fluid modeling of Landau damping
• Deep Hamiltonian networks based on symplectic integrators
• A Survey on Autonomous Vehicle Control in the Era of Mixed-Autonomy: From Physics-Based to AI-Guided Driving Policy Learning
• Stochastic stiffness identification and response estimation of Timoshenko beams via physics-informed Gaussian processes
• Highly-scalable, physics-informed GANs for learning solutions of stochastic PDEs
• Stability selection enables robust learning of partial differential equations from limited noisy data
• Variational Autoencoders for Learning Nonlinear Dynamics of Physical Systems
• Convergence rate of DeepONets for learning operators arising from advection-diffusion equations
• Data-driven Discovery of Partial Differential Equations for Multiple-Physics Electromagnetic Problem
• Solving Optical Tomography with Deep Learning
• Physics-Informed Deep Learning for Traffic State Estimation
• Computational model discovery with reinforcement learning
• Data-driven femtosecond optical soliton excitations and parameters discovery of the high-order NLSE using the PINN
• A Data Driven Method for Computing Quasipotentials
• Machine Learning for Robust Identification of Complex Nonlinear Dynamical Systems: Applications to Earth Systems Modeling
• Learning Dynamics from Noisy Measurements using Deep Learning with a Runge-Kutta Constraint
• Particles to Partial Differential Equations Parsimoniously
• Robust Data-Driven Discovery of Partial Differential Equations under Uncertainties
• FD-Net with Auxiliary Time Steps: Fast Prediction of PDEs using Hessian-Free Trust-Region Methods
• Computational Rational Engineering and Development: Synergies and Opportunities
• On the equivalence of molecular graph convolution and molecular wave function with poor basis set
• RODE-Net: Learning Ordinary Differential Equations with Randomness from Data
• HJB Optimal Feedback Control with Deep Differential Value Functions and Action Constraints
• A composable autoencoder-based iterative algorithm for accelerating numerical simulations
• Physics-informed Gaussian process model for Euler-Bernoulli beam elements
• On Theory-training Neural Networks to Infer the Solution of Highly Coupled Differential Equations
• Deep neural networks for geometric multigrid methods
• Geometry encoding for numerical simulations
• Uncertainty Quantification of Locally Nonlinear Dynamical Systems using Neural Networks
• A Neural Network Ensemble Approach to System Identification
• Multiscale and Nonlocal Learning for PDEs using Densely Connected RNNs
• Machine learning for adjoint vector in aerodynamic shape optimization
• Finite Difference Neural Networks: Fast Prediction of Partial Differential Equations
• A Latent space solver for PDE generalization
• A Temporal Kernel Approach for Deep Learning with Continuous-time Information
• Theory-training deep neural networks for an alloy solidification benchmark problem
• Differentiable Spline Approximations
• Connecting implicit and explicit large eddy simulations of two-dimensional turbulence through machine learning
• Interface Retaining Coarsening of Multiphase Flows
• Invertible Surrogate Models: Joint surrogate modelling and reconstruction of Laser-Wakefield Acceleration by invertible neural networks
• Weakly-supervised learning on Schrodinger equation
• Learning the gravitational force law and other analytic functions